Glioblastoma is the most common and malignant primary brain tumour in adults, with a dismal prognosis. This is partly due to considerable inter- and intra-tumour heterogeneity. Changes in the cellular energy-producing mitochondrial respiratory chain complex (MRC) activities are a hallmark of glioblastoma relative to the normal brain, and associate with differential survival outcomes. Targeting MRC complexes with drugs can also facilitate anti-glioblastoma activity. Whether mutations in the mitochondrial DNA (mtDNA) that encode several components of the MRC contribute to these phenomena remains underexplored. We identified a germ-line mtDNA mutation (m. 14798T > C), enriched in glioblastoma relative to healthy controls, that causes an amino acid substitution F18L within the core mtDNA-encoded cytochrome b subunit of MRC complex III. F18L is predicted to alter corresponding complex III activity, and sensitivity to complex III-targeting drugs. This could in turn alter reactive oxygen species (ROS) production, cell behaviour and, consequently, patient outcomes. Here we show that, despite a heterogeneous mitochondrial background in adult glioblastoma patient biopsy-derived cell cultures, the F18L substitution associates with alterations in individual MRC complex activities, in particular a 75% increase in MRC complex II_III activity, and a 34% reduction in CoQ10, the natural substrate for MRC complex III, levels. Downstream characterisation of an F18L-carrier revealed an 87% increase in intra-cellular ROS, an altered cellular distribution of mitochondrial-specific ROS, and a 64% increased sensitivity to clomipramine, a repurposed MRC complex III-targeting drug. In patients, F18L-carriers that received the current standard of care treatment had a poorer prognosis than non-carriers (373 days vs. 415 days, respectively). Single germ-line mitochondrial mutations could predispose individuals to differential prognoses, and sensitivity to mitochondrial targeted drugs. Thus, F18L, which is present in blood could serve as a useful non-invasive biomarker for the stratification of patients into prognostically relevant groups, one of which requires a lower dose of clomipramine to achieve clinical effect, thus minimising side-effects.

Components of the mitochondrial electron transport chain have recently gained much interest as potential therapeutic targets. Since mitochondria are essential for the supply of energy that is required for both angiogenic and tumourigenic activity, targeting the mitochondria represents a promising potential therapeutic approach for treating cancer. Here we investigate the established anti-angiogenesis drugs combretastatin A4, thalidomide, OGT 2115 and tranilast that we hypothesise are able to exert a direct anti-cancer effect in the absence of vasculature by targeting the mitochondria. Drug cytotoxicity was measured using the MTT assay. Mitochondrial function was measured in intact isolated mitochondria using polarography, fluorimetry and enzymatic assays to measure mitochondrial oxygen consumption, membrane potential and complex I–IV activities respectively. Combretastatin A4, OGT 2115 and tranilast were both shown to decrease mitochondrial oxygen consumption. OGT 2115 and tranilast decreased mitochondrial membrane potential and reduced complex I activity while combretastatin A4 and thalidomide did not. OGT 2115 inhibited mitochondrial complex II–III activity while combretastatin A4, thalidomide and tranilast did not. Combretastatin A4, thalidomide and OGT 2115 induced bi-phasic concentration-dependent increases and decreases in mitochondrial complex IV activity while tranilast had no evident effect. These data demonstrate that combretastatin A4, thalidomide, OGT 2115 and tranilast are all mitochondrial modulators. OGT 2115 and tranilast are both mitochondrial inhibitors capable of eliciting concentration-dependent reductions in cell viability by decreasing mitochondrial membrane potential and oxygen consumption.